304 H. M. BERNARD. 



and larger fields tlian the retina. The discovery of con- 

 tinuous nerve-paths through the retina along lines in which 

 no one acquainted with the current views and theories would 

 have thought of looking for them, seems to demand some 

 revision and readjustment of modern doctrines with regard to 

 sevei'al important morphological problems. And yet there is 

 not so much that is really new in this paper, for, wheu 

 critically reviewed, it will be found that nearly every fact 

 has been seen and described by other workers. What the 

 retina has now revealed is not so much new facts as new 

 clues, which will, I believe, enable us to co-ordinate a number 

 of hitherto isolated and, ou that account, often neglected 

 observations. 



The search for direct continuity between the nerve-strands 

 and the rods of the vertebrate retina, in which rods there is 

 every reason to believe the nerves must terminate, seems iu 

 recent years to have been abandoned. The reason is not 

 difficult to understand. The problem had baffled all the 

 leading retinologists of the latter half of last century, and 

 this fact prepared the way for a new theory of nerve action 

 which rendered direct continuity unnecessary. Indeed, the 

 retina became one of the chief witnesses for the intermittent 

 contact theor3^ My own researches, however, soon convinced 

 me that the phenomena on which the upholders of this latter 

 theory were relying were not to be trusted so far as the retina 

 is concerned. As soon as it dawned upon me that the 

 elaborate series of ganglionic cells with their systems of 

 dendrites, which were supposed to convey the stimulus from 

 the rods to the nerves, admitted of another find much simpler 

 interpretation, I naturally once more took up the quest for 

 the nerve-paths, without, however, any hope of solving a 

 problem which had appeared insoluble. The discovery came, 

 as it were, of itself some three or four years ago in the way 

 described in this paper. 



The descriptions of the " ganglionic cells " of the vertebrate 

 retina, as given in the text-books, are familiar to all. Asso- 

 ciated with such names as those of Corti, Koelliker, H. Miiller, 



